It is well established that people with atrial fibrillation (AF) are at increased risk of stroke. Stasis of blood in the atria (the upper chambers), which occurs when the atria fibrillate, is thought to increase the risk of clot formation, particularly within a structure called the left atrial appendage. If this clot breaks off, it can travel through the circulation to the brain, causing brain damage, ‘aka’ a stroke. Overall, this risk is 5-fold higher compared to people without AF. Although this association is extremely well documented, the exact role of AF in causing stroke is much less clear and there is increasing evidence that AF may perhaps be a marker of stroke risk rather than actually causing strokes. Nevertheless, there is compelling evidence that patients with AF who also have risk factors for stroke, benefit from taking an anticoagulant indefinitely to mitigate that risk, which far outweighs the risk of bleeding in the vast majority of patients.
Diagram showing right and left atria and left atrial appendage
How AF can lead to a stroke
We use a stroke risk factor scoring system, called the CHA2DS2-VASc score, to evaluate stroke risk in patients with AF and this is explained in detail in the ‘Atrial fibrillation & atrial flutter’ section. All patients with scores of 1 or more (in men), or 2 or more in women, should now be offered treatment with an anticoagulant and, since 2015, there have specific payments available for GPs in the UK to identify all their patients with AF, risk stratify them according to the CHA2DS2-VASc score and then offer them appropriate anticoagulant treatment.
Historically, warfarin and some other Vitamin K antagonists were the only drugs available for anticoagulation for patients with AF. Warfarin has been shown to reduce the risk of stroke overall by 70% in patients with AF, which is a very dramatic effect for a drug treatment. Unfortunately, warfarin is a very difficult drug to prescribe and requires regular blood tests to measure the ‘INR’ or International Normalised Ratio, to monitor its effect. This has traditionally been undertaken in specific ‘warfarin clinics’, but now many GPs offer this service to their patients and there is also the option of home monitoring now using ‘finger-prick’ testing.
However, the ‘blood-thinning’ effect of warfarin (and the INR) can be disturbed by the amount of Vitmain K in one’s diet (found in significant amounts in green vegetables and olive oil, amongst other foods). It can also be affected by taking or starting/stopping certain medications which interact with warfarin. Thus warfarin has significant food and drug interactions making life potentially difficult for the people who need to take it. If the INR becomes too high, the risk of bleeding increases, but if too low, the risk of stroke increases. This is known as having a narrow therapeutic window.
We can measure how often warfarin treatment is within this therapeutic window by calculating the Time in Therapeutic Range, or TTR. TTRs of 55-60% are considered good warfarin control by most measures, which is of course means the drug is ineffective for large proportions of the time. It is not surprising, based on these limitations, that alternative to warfarin have been sought for some time.
It had been clear for some time that new alternatives to warfarin were needed desperately and now four new anticoagulants have been developed over the last decade, known collectively as the Novel Oral Anticoagulants (or NOACs). Most are not novel anymore, but the name has stuck for now.
There are broadly speaking 2 classes of NOAC – ones which prevent Factor Xa working (rivaroxaban, apixaban and edoxaban) and one which works on Factor II (dabigatran). Although there are many subtle differences between them, none of them require blood-test monitoring or INR measurements, they have limited interactions with foods (including green vegetables) and other drugs and overall there is a roughly 50% lower risk of intra-cranial haemorrhage (bleeding in the brain) with these drugs than warfarin. In other words, NOACs are safe than warfarin. Rivaroxaban and edoxaban are once daily and the others are twice daily.
All the NOACs have been tested versus warfarin in large randomised trials in patients with AF and all have shown at least the same efficacy at preventing embolic stroke as warfarin (strokes due to a clot from the heart). Some were marginally better than warfarin but each trial had different patient populations and direct comparisons between the NOAC trials is not valid. Bleeding risks between differ too but, in general, all are broadly comparable with warfarin except they all have much lower rates of intra-cranial bleeding. They are all now widely used for stroke prevention in patients with AF, although the UK has been the slowest adopter of the NOACs in Europe. This is partly due to cost and partly due to the conservative prescribing habits of UK doctors, in particular GPs. Fortunately, attitudes appear to be changing at last and in 2014, NICE stipulated that all patients diagnosed with AF who meet criteria for anticoagulation according to their CHA2DS2-VASc score should be offered the choice between NOACs or warfarin after a discussion between the pro’s and con’s of each.
Based on their profile from these trials, the European Society of Cardiology now recommends NOACs should be considered in preference to warfarin for the vast majority of patients with AF.
As we do not use blood test monitoring for NOACs, it is very important they are taken every day, as prescribed, and doses not missed. You may need to stop taking it for short periods if you need surgery or other procedures. If you develop serious bleeding, it is important you seek medical attention straight away.
It is worth noting that NOACs and warfarin do not themselves cause bleeding but will potentially lengthen the time for you to stop bleeding if it occurs. The risk of major bleeding in people taking anticoagulants is low and affects approximately 1-3 in 100 people a year. If you experience a major bleed (with severe blood loss and/or symptoms requiring treatment in hospital) and you are on an anticoagulant, it can be treated successfully in approximately 90%
NOACs are only indicated for patients with what is known as non-valvular AF. This means if you have AF and also have rheumatic heart valve disease (particularly of the mitral valve) or a severely leaking mitral valve for another reason, then you should only be considered for warfarin. Similarly, if you have had heart valve surgery with a metallic valve replacement, then it is only safe for you to receive warfarin, not a NOAC.
NOACs are excreted by the kidneys, so if you have abnormal kidney function, it may not be safe for you to take a NOAC depending on your blood tests, although some NOACs are better suited to this than others.
If you are at risk of bleeding then it may not be suitable for you to take a NOAC or warfarin. In some cases, particularly if you have suffered a severe haemorrhage or bleed in the past, you may be suitable for a left atrial appendage occlusion device (please see section on left atrial appendage occlusion devices).
If you cannot take a NOAC or warfarin for AF, you should not be offered aspirin instead as this has been shown to be ineffective at reducing stroke in AF but has a similar bleeding risk. You may, of course, have been prescribed aspirin for a separate reason.
Frequent falls are not a reason to prevent you from being prescribed an anticoagulant.
The effects of warfarin last for days and so need to be reversed if serious bleeding develops. Injections of clotting factor-like substances can reverse the effect of warfarin immediately for a few hours. This agent is called Pro-thrombin complex concentrate (PCC). It should be available in all A&E departments and large hospitals in the UK or anywhere where surgery/procedures are performed. A substance called Vitamin K can also be administered which has a much longer term effect on warfarin but takes about 4 hours to start working, so both agents may be given together when bleeding occurs.
There is a lot of misinformation about reversal of NOACs. Many patients come to see me and have been told these drugs cannot be reversed or have no antidote. This is just disingenuous. Fortunately, the NOACS all have much shorter effects (or half-life) than warfarin and so wear off much more quickly. Secondly, PCC (described above) reverses the effects of the Factor Xa inhibitors rivaroxban, apixaban and edoxaban seemingly very effectively. Our confidence of this reversal effect is now so high that many centres around the world routinely, including my centre at Barts – the largest in Europe, perform AF ablation on patients taking these drugs and are not asked to stop them for the procedure. Dabigatran, the Factor II inhibitor, now has its own specific antidote which is becoming widely available and immediately reverses the effects of this drug.
It should be noted that in some cases bleeding cannot be stopped in patients on warfarin or NOACs despite using the measures above (and using some additional substances) and this will likely always remain the case. Fortunately, this risk is very small and we know from the large trials of the NOAC drugs that the risk of fatal haemorrhage on NOACs is extremely low (and this was in the era before PCC’s effects on the Xa inhibitors was known about or dabigatran’s antidote was available) and much lower than warfarin.
AF Ablation in the left atrium
There are many conditions which can increase the risk of bleeding but a few can be modified to reduce this risk, which is particularly helpful in people treated with anticoagulants. These include:
– High blood pressure. It is important this is controlled well and medicines prescribed to control are taken regularly blood pressure checked regularly
Most people should expect to take an anticoagulant indefinitely if they are prescribed it for stroke prevention in AF. This still applies even if you have had a successful ablation as we can never know whether the AF will return in the future. Exceptions to this include people who have a CHADS-VASc score of 0, as they are at very low risk of stroke and patients who have undergone successful ablation of typical atrial flutter (see Atrial fibrillation and Atrial flutter page).
Sadly, no. Current guidelines recommend anticoagulants are taken continuously even if you are currently in normal rhythm but there are studies ongoing looking at this option.